Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Flexible Group Behavior Randall W. Hill, Jr. Jonathan Gratch USC Information Sciences Institute

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Agenda Project Overview Technology Technology Transition and Application

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Project Overview Problem Statement Hypotheses Research Objectives Products

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Problem Statement Create groups of autonomous entities whose collective behavior is coherent.  Reduce cost of exercises: replace human controllers with autonomous C2 entities and intelligent forces  Provide operational realism: groups of autonomous entities must be goal-oriented, flexible, and coordinated.

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Hypotheses Flexible group behavior arises from the ability of individuals to:  Understand the behavior of groups of others  Plan missions for groups against groups  Execute a mission in coordination with others  Replan to handle situation interrupts  Interleave understanding, planning, execution, and replanning (known as Continuous Planning)  Collaborate with other others in planning

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Research Objectives Continuous Planning  Understand evolving situations  Plan dynamically and continuously  Achieve goals despite unplanned events Collaborative Behavior  Coordinate group behavior  Understand behavior of other groups  Reason about organizational constraints

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Research Objectives Situation Awareness  Current situation  Need a consolidated picture  Requires assessment at multiple echelons  Individual entities (e.g., pilots)  Commanders (e.g., company, battalion,...)  Future situation  Plan for future sensing requirements  Formulate Priority Intelligence Requirements (PIR)

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Products Company and Battalion Commander Agents  Attack Helicopter Battalion  Domain-specific knowledge bases Planning Architecture  Domain-independent  Continuous  Collaborative / Socially Aware Situation Awareness  Perceptual attention in pilot  Pattern-driven situation assessment

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Technology Research areas  Continuous Planning  Collaborative Planning  Situation Awareness  Perceptual Attention Achievements Lessons Learned Open Issues

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Research Areas Continuous Planning

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review What are Plans? Hierarchically ordered sequences of tasks Plans capture assumptions  Column movement assumes enemy contact unlikely Plans capture task dependencies  Move_to_Holding_Area results in unit being at the HA, (precondition to moving to the Battle_Position)  OPFOR and Co must be at the Engage_area simultaneously

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Continuous Planning Understand situation (sit. awareness)  Monitor state of the world  Interpret events and relate to goals and plans Generate plans  Sketch basic structure via decomposition  Fill in details with causal-link planning Execute plans  Initiate tasks with satisfied preconditions  Terminate tasks whose effects have occurred Repair Plans  Recognize situation interrupt  Repair plan by adding, retracting tasks

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Architecture of a Continuous Planner Planner (General Purpose Reasoner) Plans (Tactical) Domain Theory (Tactical) World Model Situation Assessment Synthetic Battlespace Situation Reports, Sensing Facts, inferences Expectations OPORDER Other Communications

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Plan Generation Example Destroyed(Enemy) Attack(A, Enemy) Move(A,BP) Engage(A,Enemy) at(A,BP)at(A,FARP) at(Enemy,EA) at(A,BP)Destroyed(Enemy) at(A,FARP) at(Enemy,EA) World Model... init

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Company A plan Company B plan CSS plan Move Engage Return Move OPFOR Plan Move Battalion Tactical Plans Co Deep Attack Co Deep Attack FARP Operations

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Situation Interrupts Happen! destroyed(Enemy) Attack(A, Enemy) Move(A,BP) Engage(A,Enemy) at(A,BP)at(A,FARP)at(A,BP)destroyed(Enemy) at(A,FARP) at(Enemy,EA) Current World active(A) Start of OP ADA Attack active(A)

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Reacting to Situation Interrupt Situations evolve unexpectedly  Goals change, actions fail, intelligence incorrect Determine whether plan affected  Invalidate assumptions?  Violate dependency constraints? Repair plan as needed  Retract tasks invalidated by change  Add new tasks  Re-compute dependencies

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Research Areas Collaborative Planning

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Collaborative Planning Maintain multiple plans in memory  Extend plan network to include others’ plans  Enables detection of interactions among plans Model the social stances taken toward others  Enables representation of social & organizational roles Represent (group) decision-making process  Impacts when and what one should communicate Manage the planning using social modulators  Depends on plan interactions and decision-making phase  Stance of the planner with respect to phase and role

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Plan Interaction Example Move(A,BP) Engage(A,Y) Dead(Y) Move(CSS,HQ) at(CSS,HQ) at(CSS,FAA) at(fuel,FAA) at(fuel,HQ) at(A,BP) at(A,FAA) at(A,BP) at(fuel,FAA) Operation Begins Combat Service Support Plan Attack Helicopter Company Plan resupplied(HQ)

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Social Stances Authoritative  Order subordinate to alter his plans Deferential  Change my plans to de-conflict with superior Helpful  Help peer to resolve conflicts in plan Self-serving / Rude Adversarial  Try to introduce conflict in other agent’s plan

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Social Stance: Being Helpful Planning issues  Propose tasks that facilitate others’ plans  Avoid introducing threats into others’ plans Communication Issues  Collaboration protocols: propose, accept, counter  Relevance reasoning  Which of my tasks would others want to know: e.g. “Honey, I’m going to the market”

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Social Stance: Self-Serving Planning issues  Notice things that others might do for me  Ignore threats I introduce into other’s plans Unless that keeps them from doing things for me Communication Issues  Deception  e.g. Someone might not help me if the knew what I was really planning

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Model Decision-Making Process MDMP Mission Analysis Generate Plan Analyze Plan Transmit Plan Execute Plan Brief Subordinates Merge Subplans Repair Plan Transmit Changes Situation Interrupts (Military Decision- Making Process)

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Modulating the Planning Must model when to use different stances  Involves organizational issues Where do I fit in the organization  Stances may change E.g., During COA Analysis, adopt an adversarial stance towards one’s own plans Must model how stances influence planning  How do we alter COA generation

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Planner Architecture Planner (General Purpose Reasoner) Plans (Tactical Operations) Domain Theory (Tactical Operations) World Model Situation Assessment Synthetic Battlespace Situation Reports, Sensing Facts, inferences Expectations OPORDER Other Communications Plan Manager Management Theory (Domain Independent) Management Plans

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review When To Use A Stance Model the collaborative planning process  Includes management tasks that modulate the generation of tactical plans  Tasks refer to specific tactical plans  Specify preconditions on changing stance  Includes knowledge of one’s organizational role Planner constructs management plans  Use same mechanisms as tactical planning

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Management Plan Example Explicitly model the Military Decision Making Process COA Development Authoritative towards subordinates Deferential towards superiors Adversarial towards OPFOR COA Analysis Authoritative towards OPFOR Adversarial towards self (war gaming) TasksStances

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Implementing Stances Implemented as search control on planner  Plan manager Takes executing management tasks Generates search control recommendations Example: Deferential Stance  When giving orders to subordinates Indicate subset of plan is fixed ( defer to this ) Indicate rest of plan is flexible  Plan manager enforces these restrictions

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Interaction Example Move(A,BP) Move(CSS,HQ) at(CSS,HQ) at(CSS,FAA) at(gas,FAA) at(gas,HQ) at(A,BP) at(A,FAA) at(gas,FAA) Initial State Planner Retract Deferential towards Combat Service Support Plan Make CSS Planner defer to Company A’s Plan Manager

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Planner Architecture Planner (General Purpose Reasoner) Plans (Tactical Operations) Domain Theory (Tactical Operations) World Model Situation Assessment Synthetic Battlespace Situation Reports, Sensing Facts, inferences Expectations OPORDER Other Communications Plan Manager Management Theory (Domain Independent) Management Plans

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Research Areas Situation Awareness

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Situation Awareness Planner needs a consolidated picture of the current situation in the battlespace  Determines which goals and tasks are achievable  Influences the choice of strategies and actions  Allows the detection of imminent plan failure  Enables re-planning Situation assessment updates the World Model  Monitor plans with respect to world model  Situation awareness = world model + plans/tasks

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Situation Awareness Performed at multiple echelons  Scouts performing reconnaissance of battlespace  C2 staff assimilates scouting and sensor reports General process:  Identify entities  Classify groups of entities as units  Determine units’ functionality, capabilities, plans, intent Technical Issues  Pilot awareness and information overload - See section on human behavior representation  Situation assessment techniques

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Situation Awareness: Higher Echelons Command Entity Situation Reports

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Situation Assessment Identify entities  Fuse scouting reports Classify groups of entities as units  Cluster entities into unit-sized groups  Classify units into functional types Determine capabilities, plans, intent

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review What Situation Assessment Involves Functional level  Determine the functions of entities and units  Determine possible sequence of actions Structural level  Identify different types of entities  Group entities into functional units  Create hierarchical structures of units

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Approach to Situation Assessment Functional level  A database of templates of units, their spatial relationships, functions, capacities, deployment conditions, etc. Structural level  A hierarchical encoding (pattern) of units and their spatial relationships  Templates of units  Perceived entities  Pattern matching algorithms  Match the pattern of perceived entities against patterns of templates  Identify the most relevant templates

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Future Situation Awareness How tactical intelligence influences planning Future situation: knowledge goals  What will I need to know for this plan to work?  Establish Priority Intelligence Requirements (PIR)  What commander needs to know about opposing force  Drives the placement of sensors and observation posts  Constrains the pace of plan execution Rarely addressed in current C 2 models

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Intelligence Critical for Realistic C 2 Interplay between intelligence and COA Development  Intelligence guides COA development  COA development drives intelligence needs  Intelligence availability constrains actions  Some COA’s may be abandoned if inadequate intelligence

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Intelligence Critical for Realistic C 2 Intelligence imposes temporal constraints  When can a satellite observe?  How long to insert surveillance (LRSU)?  How long before I must commit to COA?

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Intelligence critical for realistic C 2 Intelligence collection must be focused  Commanders must:  Prioritize their intelligence needs  Understand higher-level intelligence priorities  Provide intelligence guidance to subordinates  e.g. Simulation Information Filtering Tool [Stone et. al]

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Brigade Planning (simplified)  Identify Engagement Area (EA Pad) Should canalize OPFOR and restrict movement  Identify launch time Require 2-hour notice EA Pad AA Lincoln Attack 2 nd echelon tank division (TD)

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review PL ECHO Brigade PIR  When will TD leave AA Lincoln? Verifies enemy intent  When will TD reach PL Echo? Satisfies the need for 2-hour notice Further verifies enemy intent Location of PL Echo driven by PIR EA Pad AA Lincoln 2hrs

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review EA Pad PL ECHO Intelligence Plan SLAR Monitor movement from assembly area LRSU Trigger attack: TD 2hrs from EA Pad Assembly Area

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Final Brigade Plan Execute Mission Arrive at EA Break Contact Decision Point H H+2 H+3H-8 H-10 Insert LRSULRSU monitor PL Echo Deep Attack SLAR monitor AA

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Automating PIR Identify PIR in my own plans  Find preconditions, assumptions, and triggering conditions that are dependent on OPFOR behavior Extract PIR from higher echelon orders  Specialize as appropriate for my areas of operation Derive tasks for satisfying PIR  Sensor placement Ensure consistency of augmented plans

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Identifying PIR Examine COA dependencies on OPFOR  e.g. Precondition of engaging: OPFOR will-be-at EA Pad at time H+2 Look for dependencies that:  Are not under my direct control  Are uncertain Implemented with PIR recognition schema:  Abstract rules that scan plans and assert PIR  Some domain-independent, some domain-specific

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Interpreting Higher Level Guidance Need to convert into PIR at my echelon  e.g. Brigade’s PIR:  When will lead regiment reach forward defense becomes Battalion PIR  When will lead battalion of lead regiment reach forward defenses Implemented by specialization rules  Encode doctrinal and terrain relationships

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Deriving Sensor Plans Implemented via tactical planning mechanism  PIR represented as “knowledge goals”  Domain theory augmented with sensing tasks  Sensing tasks achieve knowledge goals  Tasks encode maneuver / temporal dependencies  Planning process fills in details  Sensing tasks added to achieve knowledge goals  e.g. Observe TD activity near PL_ECHO  Other tasks added to satisfy maneuver dependencies  e.g. Use UH-60 to insert LRSU near PL_ECHO

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Ensuring Consistency Implemented via tactical planning mechanism  If PIR goals cannot be satisfied, COA is invalid or Use unsatisfied PIR to request external assets Sensing plans constrain timing of events  If temporal constraints inconsistent, COA is invalid

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Research Areas Perceptual Attention

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Pilot Situation Awareness Synthetic worlds are information rich  100’s of other entities  Vehicle instruments  Terrain, weather, buildings, etc.  Communications (messages)  Amount of information will continue to increase …. Perceive, understand, decide and act  Comprehend dynamic, complex situations  Decide what to do next  Do it!

Information Overload

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Roots of the Problem Naïve vision model  Entity-level resolution only  Unrealistic field of view (360 o, 7 km radius) Perceptual-Cognitive imbalance  Too much perceptual processing  Cognitive system needs inputs, but …  It also needs time to respond to world events

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program ReviewApproach Create a focus of attention  Apply attention mechanisms to entity perception initially  Incorporate filters  Implement a zoom lens model (covert attention) Stages of perceptual processing  Attention in different stages: preattentive & attentive Control the focus of attention  Goal-driven  Stimulus-driven

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Zoom Lens Model of Attention (Eriksen & Yeh, 1985) Attention limited in scope  Multi-resolution focus  Magnification inversely proportional to field of view Low resolution  Large region, encompassing more objects, fewer details  Perceive groups of entities as a coherent whole High resolution  Small region, fewer objects, more details  Perceive individual entities (e.g., tank, truck, soldier)

Low Resolution

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Perceptual Grouping Preattentive Gestalt grouping  Involuntary  Proximity-based  Other features Dynamic Voluntary grouping K K K

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Group Features Quantity and composition Activity  Moving  Shooting Location  Center-of-mass  Bounding-box Geometric relationships with respect to pilot  Slant-range, azimuth, etc.

High Resolution

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Entity Features Location (GCS) Speed Velocity Orientation Slant Range Force Object, Object Type Vehicle Class Function Sense Name Altitude Angle Off Target Aspect Magnetic bearing Heading Status Lateral Range Lateral Separation Closing Velocity Vertical Separation

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Control of Attention Goal-driven control  Agent controls the focus / resolution of attention  Low resolution: Scouting groups of enemy; escorting group  High resolution: Search for air-defense entities; engage target  Sets filters that select entities for WM Stimulus-driven control  Attention can be captured involuntarily by a visual event  Muzzle flash (luminance contrast, abrupt onset)  Sudden motion (abrupt onset)

Sea Land Overwatch Position Overwatch Position Transport Carrier Escort Carrier Rendezvous Point Escort task Orient on group Voluntary grouping Goal-driven Attention

Low Resolution High Resolution Stimulus-driven Attention

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Technology Research areas Achievements Lessons Learned Open Issues

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Achievements General agent architecture  Continuous Planning  Collaborative Planning  Behavioral Moderators (ARI)  Perceptual Attention Model Visualization Tools  Immersadesk demo Interoperability with other virtual worlds  Integration with tutoring environment  (Jack & Steve)

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review B Company A Company Trigger Attack Here Battalion Assembly Area Artillery Support Here

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Technology Research areas Achievements Lessons Learned Open Issues

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Lessons Learned Intelligent, reactive behavior is not sufficient  Need ability to plan Planning is not sufficient  Need ability to monitor plan execution  Need situation awareness  Need ability to replan Intelligent, continuous planning is not sufficient  Need to function in an organization Divide and Conquer approach does not work  Situation awareness, planning, execution, monitoring, replanning, collaboration are highly interdependent

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Technology Research areas Achievements Lessons Learned Open Issues

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Open Issues Need to improve situation awareness  Fusion  Situation assessment  Plan recognition Need richer organizational models  “Teamwork” protocols  Different organizational types Need generic cross-echelon domain libraries  Multi-echelon mission-to-task decomposition

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Technology Transition and Application

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Operational Fit Command and Control Nodes in Simulation  Army: Command entities (Co, Bn)  Air Force: Airborne command elements (ACE)  Navy: Air controllers? High-fidelity Models of Human Behavior  Tutoring systems  Interactive Learning Environments  Asymmetric threat

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review How Does This Make The World Better? Better Training  Larger-scale exercises  Greater operational realism  Institute for Creative Technologies (USC-Army) Reduce Cost of Exercises  Autonomy requires fewer controllers  C2 modeling supports higher-echelon simulations Contributes to interactive education and entertainment  Interactive stories in pedagogical environments

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Domains of Applicability Interactive Story Drive Systems Immersive Training Applications  Synthetic Battlespaces  Emergency Management Systems (Civilian) Mission rehearsal Wargaming

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Efforts Required to Operationalize Create “hardened” implementation  C, C++ Knowledge Acquisition Tools  Improve domain representation language  Possibly integrate with EXPECT project Create HLA compliant version Broaden class of domain models  Organize knowledge into libraries

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Lessons Learned Products Lessons Learned Products Need ability to plan Need ability to monitor plan execution Need situation awareness Need ability to replan Need to function in an organization Situation awareness, planning, execution, monitoring, replanning, collaboration are highly interdependent Planning Architecture  Domain-independent  Continuous  Situation Awareness  Collaborative / Socially Aware Company and Battalion Commander Agents  Attack Helicopter Battalion  Domain-specific knowledge bases

Flexible Group Behavior USC/ISI 1-5 November 1999 Synthetic Forces Advanced Simulation Technology Thrust Final Program Review Who Should Be Interested? JSIMS  WARSIM  NASM Institute for Creative Technologies  USC-ICT: U.S. Army Research Center Real World C3I Wargaming applications Immersive, interactive education and entertainment applications